Datasets:

THULab
/

sam_fold_cloth_merge

A bimanual cloth-folding manipulation dataset recorded on a SAM dual-arm robot (robot_type = sam_bimanual). Each episode is a teleoperated demonstration of the single task "Fold a cloth", capturing the synchronized joint states and commanded actions of both arms.

Episodes: 108 demonstrations of one task ("Fold a cloth").
Frames: 112,536 timesteps total, recorded at 30 fps.
Robot: SAM bimanual (two 7-DoF arms — 6 joints + gripper per arm).
Source format: LeRobot v2.1.

Schema (TsFile structure)

Time (INT64, milliseconds) — per-frame timestamp, restarts at 0 for each episode (round(timestamp * 1000); ~33 ms apart at 30 fps).
episode_index (TAG) — demonstration id (0–107). Query a single episode with WHERE episode_index=0.
task_index (TAG) — task id (always 0; the dataset has one task).
frame_index (FIELD, INT64) — frame number within the episode.
sample_index (FIELD, INT64) — the source global index column.
action_0 .. action_13 (FIELD, FLOAT) — commanded joint targets.
observation_state_0 .. observation_state_13 (FIELD, FLOAT) — measured joint states.

The 14 action / state channels are the bimanual joint set, in order: left_shoulder_pan, left_shoulder_lift, left_elbow_flex, left_wrist_flex, left_wrist_pan, left_wrist_roll, left_gripper, right_shoulder_pan, right_shoulder_lift, right_elbow_flex, right_wrist_flex, right_wrist_pan, right_wrist_roll, right_gripper.

The source timestamp column is dropped because it equals Time / 1000. Camera videos are not included in this repository (see below).

Usage

Read the .tsfile file with the Apache TsFile Java or Python SDK.

# pip install tsfile
from tsfile import TsFileReader
reader = TsFileReader("data/sam_fold_cloth_merge.tsfile")
# query one episode: WHERE episode_index=0

Videos

The original dataset records four camera streams per episode (observation.images.top_camera, bottom_camera, left, right). These MP4 videos are not included here — they are not time-series data. Retrieve them from the original dataset under girardijp/sam_fold_cloth_merge/videos.

Source & license

Original dataset: https://huggingface.co/datasets/girardijp/sam_fold_cloth_merge
Author / publisher: girardijp
License: not declared by the original dataset; please defer to the original.

Downloads last month: 81

Total file size:

2.94 MB